Neisseria and Lipo-oligosaccharide (LOS)
While Neisseria species commonly colonize many mammalian hosts, human beings are the only species subject to invasive disease by members of this species. Neisseria meningitidis is the etiologic agent for septicemia and meningitis that may occur in epidemic form. Neisseria gonorrhoeae is the causative agent of gonorrhea and its manifold complications. These organisms, particularly the gonococcus, have proved remarkably adept at varying the antigenic array of their surface-exposed molecules, notably their adhesive pili and opacity-related (opa) proteins. The genetic mechanisms for the variation of pilus (Meyer et al., 1982, Cell 30:45; Haas and Meyer, 1986, Cell 44:107; Koomey et al., 1987, Genetics 117:391; Swanson and Koomey, 1989, American Society for Microbiology, Washington, 743-761) and opa protein (Stern et al., 1986, Cell 47:61; Meyer et al., 1990, Ann. Rev. Microbiol. 44:451; Bhat et al., 1991, Molec. Microbiol. 5:1889) expression are in the main well understood. Like other Gram-negative bacteria the Neisseria ssp. carry LPS in the external leaflet of their outer membranes (Johnston and Gotschlich, 1974, J. Bacteriol. 119;250). In contrast to the high molecular weight LPS molecules with repeating O-chains seen in many enteric bacteria, the LPS of Neisseria ssp. is of modest size and therefore is often referred to as lipooligosaccharide or LOS. Although the molecular size of the LOS is similar to that seen in rough LPS mutants of Salmonella ssp., this substance has considerable antigenic diversity. In the case of the meningococcus, a serological typing scheme has been developed that separates strains into 12 immunotypes (Zollinger and Mandrell, 1977, Infect. Immun. 18:424; Zollinger and Mandrell, 1980, Infect. Immun. 28:451). A remarkably complete understanding of the structure of meningococcal LPS (recently reviewed (Verheul et al., 1993, Microbiol. Rev. 57:34) has resulted from the studies of Jennings and his colleagues (Jennings et al., 1983, Carbohyd. Res. 121:233; Michon et al., 1990, J. Biol. Chem. 265:7243; Gamian et al., 1992, J. Biol. Chem. 267:922; Pavliak et al., 1993, J. Biol. Chem. 268:14146). In the case of Neisseria gonorrhoeae, antigenic variability is so pronounced that a serological classification scheme has proved elusive. In part this is due to the heterogeneity of LOS synthesized by a particular strain; LOS preparations frequently contain several closely spaced bands by SDS-PAGE (Mandrell et al., 1986, Infect. Immun. 54:63). Further, studies using monoclonal antibodies indicate, that gonococci are able to change the serological characteristics of the LOS they express and that this antigenic variation occurs at a frequency of 10−2 to 10−3, indicating that some genetic mechanism must exist to achieve these high frequency variations (Schneider et al., 1988, Infect. Immun. 56:942; Apicella et al., 1987, Infect. Immun. 55:1755). Because of the molecular heterogeneity and antigenic variation of the LOS produced by gonococci the determination of the structural chemistry of this antigen has proved to be a difficult problem, and definitive information based on very sophisticated analyses has only recently become available (Yamasaki et al, 1991, Biochemistry 30:10566; Kerwood et al., 1992, Biochemistry 31:12760; John et al., 1991, J. Biol. Chem. 266:19303; Gibson et al., 1993, J. Bacteriol. 175:2702). These are summarized in FIG. 1. Of particular interest is the presence of the tetrasaccharide Galβ1→4GlcNAcβ1→3Galβ1→4Glcβ1→4, which is a perfect mimic of lacto-N-neotetraose of the sphingolipid paragloboside (Mandrell et al., 1988, J. Exp. Med. 168:107; Tsai and Civin, 1991, Infect. Immun. 59:3604). In LOS this tetrasaccharide frequently bears an additional N-acetyl galactosamine residue (GalNAcβ1→4Galβ1→4GlcNAcβ1→4Glcβ1→4), and then mimics gangliosides. In some strains of gonococci an alternative side chain is found which has the structure Galα1→4Galβ1→4Glcβ1→4Hep→R (John et al., 1991, J. Biol. Chem. 266:19303). This is a mimic of the saccharide portion of globo-glycolipids (Mandrell, 1992, Infect. Immun. 60:3017), and is the structure characteristically found in Neisseria meningitidis immunotype L1.
The LOS molecules have a number of biological activities. They are potent endotoxic molecules believed to be the toxin responsible for adrenal cortical necrosis seen in severe meningococcal disease. They serve as the target antigen for much of the bactericidal activity present in normal or convalescent human sera (Rice et al., 1980, J. Immunol. 124:2105). Gonococci possess a very unusual sialyl transferase activity which is able to use externally supplied CMP-NANA and add N-acetyl neuraminic acid to the LOS on the surface of the organism (Nairn et al., 1988, J. Gen. Microbiol. 134:3295; Parsons et al., 1989, Microb. Pathog. 7:63; Mandrell et al., 1990, J. Exp. Med. 171:1649). Group B and C meningococci, have the capacity to synthesize CMP-NANA, and frequently sialylate their LOS without requiring exogenous CMP-NANA (Mandrell et al., 1991, J. Bacteriol. 173:2823). In Neisseria meningitidis strain 6275 immunotype L3, the sialic acid unit is linked α2→3 to the terminal Gal residue of the lacto-N-neotetraose (Yamasaki et al., 1993, J. Bacteriol. 175:4565). The levels of CMP-NANA found in various host environments is sufficient to support this reaction (Apicella et al. 1990, J. Infect. Dis. 162:506). The sialylation of the LOS causes gonococci to become resistant to the antibody-complement dependent bactericidal effect of serum (Parsons et al., 1989, Microb. Pathog. 7:63). The resistance is not only to the bactericidal effect mediated by antibodies to LOS, but to other surface antigens as well (Wetzler et al., 1992, Infect. Immun. 60:39). van Putten has demonstrated that exposure of gonococci to CMP-NANA markedly reduces their ability to invade epithelial cells in tissue culture (Van Putten, 1993, EMBO J. 12:4043). These findings strongly suggest that the ability of gonococci to vary the chemical nature of the LOS provides them with the ability to cope with different host environments (Mandrell and Apicella, 1993, Immunobiology 187:382).
Perhaps most telling, it has been found that LOS variation is selected in vivo in infections of human beings. A well characterized gonococcal laboratory strain MS11mk variant A was used to inoculate volunteers (Swanson et al., 1988, J. Exp. Med. 168:2121). In the two infected individuals over a period of 4 to 6 days the population of gonococci recovered in their urine increasingly shifted to two variants that expressed antigenically different LOS (Schneider et al., 1991, J. Exp. Med. 174:1601). A structural analysis revealed that the inoculated variant A produced a truncated LOS containing only the β-lactosyl group linked to Hep1, while one of the new variants (variant C) produced a complete LOS (Kerwood et al., 1992, Biochemistry 31:12760). This suggests that the addition of the additional sugars GalNAcβ1→3Galβ1→4GlcNAcβ1→3 is likely to be under control of a phase variation mechanism.
Little information on the genetics of LOS synthesis in Neisseria is available. A major advance has been the creation (Dudas and Apicella, 1988, Infect. Immun. 56:499) and biochemical characterization (John et al., 1991, J. Biol. Chem. 266:19303) of five pyocin mutants of gonococcal strain 1291, dubbed 1291a-e. Immunological and biochemical data have shown that 1291a, 1291c, 1291d and 1291e produce LOS with sequential shortening of the lacto-N-neotetraose chain, with mutant 1291e lacking the glucose substitution on the heptose. Mutant 1291b synthesizes the alternative LOS structure Galα1→4Galβ1→4Glc (see FIG. 1). Only the genetic basis of the 1291e mutant is now defined. It is a mutation of phosphoglucomutase (pgm), which precludes the synthesis of UDP-glucose, and hence the addition of the first residue of the lacto-N-neotetraose unit (Zhou et al., 1994, J. Biol. Chem. 269:11162; Sandlin and Stein, 1994, J. Bacteriol. 176:2930). It also has been shown that galE mutants of meningococcus or gonococcus produce truncated LOS in keeping with the inability to synthesize UDP-galactose (Robertson et al., 1993, Molec. Microbiol. 8:891; Jennings et al., 1993, Molec. Microbiol. 10:361).
Biosynthesis of Oligosaccharides
Oligosaccharides are polymers of varying number of residues, linkages, and subunits. The basic subunit is a carbohydrate monosaccharide or sugar, such as mannose, glucose, galactose, N-acetylglucosamine, N-acetylgalactosamine, and the like. The number of different possible stereoisomeric oligosaccharide chains is enormous.
Oligosaccharides and polysaccharides play an important role in protein function and activity, by serving as half-life modulators, and, in some instances, by providing structure. As pointed out above, oligosaccharides are critical to the antigenic variability, and hence immune evasion, of Neisseria, especially gonococcus.
Numerous classical techniques for the synthesis of carbohydrates have been developed, but these techniques suffer the difficulty of requiring selective protection and deprotection. Organic synthesis of oligosaccharides is further hampered by the lability of may glycosidic bonds, difficulties in achieving regio-selective sugar coupling, and generally low synthetic yields. In short, unlike the experience with peptide synthesis, traditional synthetic organic chemistry cannot provide for quantitative, reliable synthesis of even fairly simple oligosaccharides.
Recent advances in oligosaccharide synthesis have occurred with the isolation of glycosyltransferases. These enzymes can be used in vitro to prepare oligosaccharides and polysaccharides (see, e.g., Roth, U.S. Pat. No. 5,180,674, issued Jan. 19, 1993). The advantage of biosynthesis with glycosyltransferases is that the glycosidic linkages formed by enzymes are highly stereo and regio-specific. However, each enzyme catalyzes linkage of specific sugar residues to other specific acceptor molecules, e.g., an oligosaccharide or lipid. Thus, synthesis of a desired oligosaccharide may be limited by the availability of glycosyltransferases (see, Roth, International Patent Publication No. WO 93/13198, published Jul. 8, 1993).
Another drawback of biosynthesis is that the glycosyltransferases themselves are usually present in fairly low quantities in cells. It is difficult to obtain enough of the enzyme to be commercially practicable.
Thus, there is a great need in the art for glycosyltransferases. There is a further need for genes encoding such glycosyltransferases, to provide an unlimited source of glycosyltransferases through recombinant technology.
The citation of any reference herein should not be construed as an admission that such reference is available as prior art to the instant invention.